From four tools to one dispatch workspace
This case study shows how I redesigned the daily workspace for enterprise fleet dispatchers who were juggling four legacy tools plus Excel for every decision.
Working within a 15-year-old backend and a six-week deadline, I led the UX strategy to create a single, map-centric dashboard that cuts context switching, reduces time-to-dispatch, and gives dispatchers clear, trustworthy data instead of noisy screens.
Enterprise Fleet Management
Consolidating 4 legacy tools into a single decision surface without a backend rewrite.
Lead UX & Strategy • Enterprise • Logistics
The Problem (The Mental Model Gap)
The "Swivel Chair" Workflow
The dispatch team was routing trucks using four separate legacy tools plus Excel. Every decision meant jumping across windows, manually reconciling data, and holding everything in working memory. The friction wasn’t just in the clicks—it was in the Cognitive Load.
​
-
The 12% error rate wasn't a skill issue; it was a system design issue. Users were spending 80% of their time gathering data and only 20% deciding.
​​​​
Workflow comparison.Caption: Moving from fragmented legacy windows (left) to a unified decision surface (right).
Solution Finding
Designing for Trust in a Slow System
One of the critical challenges was a 15-year-old backend API with a 3–5 minute latency. Product leadership wanted "real-time" map animations, but Engineering confirmed that showing live dots on stale data would lead to routing errors. The Pivot: I reframed the goal from "Real-Time" to "Operational Trust." I designed a component that explicitly tells the dispatcher how reliable the data is at any moment.
The Data Freshness component turned a technical constraint (API lag) into a transparent feature, allowing dispatchers to make safer decisions.
The Evolution
Validating the Mental Model
​
Low-Fidelity: Validating the 3-column structure to kill pop-up windows.
.png)
Mid-Fidelity: Moving filters to a control bar after discovering visibility issues on low-contrast monitors.
High-Fidelity: The final empty state, guiding the user to start a prioritized task.
High-Fidelity: The final empty state, guiding the user to start a prioritized task.
​The Single Pane of Glass​
-
Actionable Queue: Work is prioritized by risk (Late/At Risk) rather than just ID.
-
Shared Context: Weather and traffic overlays are integrated directly into the map, removing the need for external browser tabs.
-
Decision Clarity: Constraints and "ETA vs SLA" are calculated instantly, reducing math errors.
By aligning the interface with the user's mental model, we achieved significant operational gains without a costly backend rewrite.
-
30% Faster: Average time-to-dispatch dropped significantly in the pilot group.
-
Errors Crushed: Routing errors fell from ~12% to ~2%.
-
Adoption: 95% of tasks moved to the new dashboard in Week 1.
.png)
